Hi Chris! Thanks again for your time, it’s truly an honour! Firstly, could you tell me a bit about your role at the NZ Herald? What exactly does data journalism entail? How does it diverge, in process, media cycle/timeline, form, content, etc from ‘traditional’ journalism?
My role at the NZ Herald is a mixture of work on large data-driven interactives, e.g. mapping burglaries at a neighbourhood level, and collaborative work with other journalists to help them make more use of data in their stories.
Data journalism is very loosely defined. I think of it as journalism where one, or all, of the sources is a data set of some description, and the job of a data journalist is to understand that something and then tell a story about it. But it also often involves bringing a different, maybe a more statistical mindset to storytelling. And many data journalists, but certainly not all, do quite a bit of programming and analysis in support of stories.
Programming, doing data analysis, and creating interactives generally takes a bit (or a lot) longer than ‘traditional’ journalism. Although, the flip side is that once an ‘interactive story’ has been built, then it can be quite quick to repeat it when new data is available.
What does your current media cycle look like? Are there any major events or dates that you typically plan or look out for content?
The big events in a data journalist’s calendar are generally elections, budget day, census release and big sporting events (which for us are primarily the Rugby World Cup and the Olympics). So, with the looming US election, and the recent release of the latest New Zealand census, it’s a relatively busy time, but it’s much quieter than during the leadup to a New Zealand election.
For these big events, I generally set up as much as possible in advance so that as soon as the data is available, I can plug it in and publish. It doesn’t always work like this. A few weeks ago, I battled with the census release data for quite a few hours after it was released before finally being able to publish.
There are also almost daily releases of official data (usually from government departments) so part of my role is looking at the upcoming releases and deciding if our readers will be interested in them. For example, there is considerable interest in inflation figures currently and very little interest in livestock slaughtering statistics.
How has your career thus far influenced/prepared you for your current role? Have you always wanted to work in the media, and how has reportage and mis/information been affected by an increasingly data driven world?
I originally trained as a scientist. A scientific background and mindset certainly helps in data journalism, but I left science relatively quickly and have spent most of my career since working on different ways to help people understand complex datasets. This led quite naturally into data journalism. I discovered data journalism in 2011 when I heard Amanda Cox give a talk about data journalism at the New York Times. It was almost a conversion moment–I left knowing that I wanted to move into data journalism–but didn’t really know how to get there.
The NZ Herald has always recruited into its data journalism roles from both journalists who want to move into data or data people who want to move into journalism. A position came up in early 2017 so I applied and got the job, which I still love.
One of my favourite definitions of data journalism comes from the London-based Bureau of Investigative Journalism: “Data journalism is simply journalism. The former is a new and trendy term but ultimately, it is just a way of describing journalism in the modern world.”
The more data-driven the world becomes, the truer this statement becomes. We need more and more journalists who are comfortable with data, because so much of our current society rests on and is mediated by data.
Something that I find puzzling is that as more and more data has become available people seem to have become less and less sceptical about that data. I feel like it wasn’t very long ago that putting a chart alongside an argument would automatically trigger a “Lies, damn lies, and statistics” type of dismissal. But now charts and graphs often seem to confer a mantle of authority, which may not be deserved. There is more data out there than ever before, but I’m not convinced there is more good data out there. One purpose of data-journalism, especially interactives, is to help readers engage with and understand important datasets. For example, an interactive map of median income by neighbourhood is much more informative, and probably more engaging, than a chart showing median income nationally.
I also want to ask what your thoughts are on AI. What role do you think it will play in the broader news/media landscape? Have you used AI in your work and do you think that it provides more benefits than detractions?
I think AI will continue, and probably accelerate, the general trend of doing more with less that began with ‘computer-assisted journalism’. There’s no shortage of time-consuming and difficult tasks in data journalism that AI could help with. ChatGPT is pretty good at explaining how to transform data from one format to another.
AI is getting quite good at extracting information from loosely structured documents like press releases. Things like ‘here are 500 police press releases, please tell me the date of each incident, the type of incident and the names of anyone who has been arrested’. Recently, it’s gotten very good with dates and incident types, but it still isn’t there on names. AI also seems to be very bad on numbers. Frequently, if I ask AI to extract numerical information from documents, it just makes stuff up.
So, on balance I’d say that for me, AI still isn’t a net benefit. We can’t afford to get things like the names wrong, so I still need to check everything. However, I know that AI has been massively helpful for some of the younger journalists who have been using it to teach them how to get started with programming and data analysis.
How have you seen the industry evolve, particularly in the advent of social media? Has the emergence of these platforms impacted data journalism, and how have you adapted to these changes?
When I started in data journalism in 2017 there seemed to be much more emphasis on large multi-layered interactives. Whereas now those seem to be the exception rather than the rule. I’m sure that at least part of this transition has been driven by the need to make journalism available on as many social media platforms as possible, a requirement that inevitably means you need to simplify what you’re doing.
Although, I’m also convinced that we wouldn’t have the large audiences we do have for the big interactives that we do, if our audiences weren’t used to spending lots of time interacting with content and ideas on social media.
The biggest shift I have seen in data journalism was caused by Covid-19. Prior to the pandemic, while many news organisations saw the importance of data journalism, it was often the province of slower, more investigative style journalism and wasn’t part of live news offerings. Covid-19 case reporting changed all that. Suddenly, rapid, accurate and detailed reporting of data became pretty much the only news. Case numbers and vaccine coverage no longer feature on homepages, but many newsrooms have continued to expect that data is part of live news coverage.
And finally, have you received a memorable pitch lately? What should people avoid when pitching to you or your team?
I think it’s fair to say the art of pitching to data journalists isn’t something the industry has really understood yet.
Many of the pitches I get are based on the misunderstanding that I want to write stories about data when I actually want to write stories from data. I’d love to get a pitch along the lines of: “Our client has collected this amazing dataset and we would like to give you one-off access to an anonymised extract of the data to see if there are any stories you could tell from the data.”
Share
Top